Search Results for author: Ayman Boustati

Found 7 papers, 1 papers with code

eCat: An End-to-End Model for Multi-Speaker TTS & Many-to-Many Fine-Grained Prosody Transfer

no code implementations20 Jun 2023 Ammar Abbas, Sri Karlapati, Bastian Schnell, Penny Karanasou, Marcel Granero Moya, Amith Nagaraj, Ayman Boustati, Nicole Peinelt, Alexis Moinet, Thomas Drugman

We show that eCat statistically significantly reduces the gap in naturalness between CopyCat2 and human recordings by an average of 46. 7% across 2 languages, 3 locales, and 7 speakers, along with better target-speaker similarity in FPT.

Transfer learning with causal counterfactual reasoning in Decision Transformers

no code implementations27 Oct 2021 Ayman Boustati, Hana Chockler, Daniel C. McNamee

In this study, we apply causal reasoning in the offline reinforcement learning setting to transfer a learned policy to new environments.

counterfactual Counterfactual Reasoning +3

VarGrad: A Low-Variance Gradient Estimator for Variational Inference

1 code implementation NeurIPS 2020 Lorenz Richter, Ayman Boustati, Nikolas Nüsken, Francisco J. R. Ruiz, Ömer Deniz Akyildiz

We analyse the properties of an unbiased gradient estimator of the ELBO for variational inference, based on the score function method with leave-one-out control variates.

Variational Inference

Amortized variance reduction for doubly stochastic objectives

no code implementations9 Mar 2020 Ayman Boustati, Sattar Vakili, James Hensman, ST John

Approximate inference in complex probabilistic models such as deep Gaussian processes requires the optimisation of doubly stochastic objective functions.

Gaussian Processes

Generalized Bayesian Filtering via Sequential Monte Carlo

no code implementations23 Feb 2020 Ayman Boustati, Ömer Deniz Akyildiz, Theodoros Damoulas, Adam M. Johansen

We introduce a framework for inference in general state-space hidden Markov models (HMMs) under likelihood misspecification.

Bayesian Inference Object Tracking

Non-linear Multitask Learning with Deep Gaussian Processes

no code implementations29 May 2019 Ayman Boustati, Theodoros Damoulas, Richard S. Savage

We present a multi-task learning formulation for Deep Gaussian processes (DGPs), through non-linear mixtures of latent processes.

Benchmarking Gaussian Processes +1

Cannot find the paper you are looking for? You can Submit a new open access paper.